Data imbalance is a common problem in machine learning and image processing. The lack of training data for the rarest classes can lead to worse learning ability and negatively affect the quality of segmentation. In this paper, we focus on the problem of data balancing for the task of image segmentation. We review major trends in handling unbalanced data and propose a new method for data balancing, based on Distance Transform. This method is designed for using in segmentation convolutional neural networks (CNNs), but it is universal and can be used with any patch-based segmentation machine learning model. The evaluation of the proposed data balancing method is performed on two datasets. The first is medical dataset LiTS, containing CT images of liver with tumor abnormalities. The second one is a geological dataset, containing of photographs of polished sections of different ores. The proposed algorithm enhances the data balance between classes and improves the overall performance of CNN model.
Study question To compare the clinical effectiveness of ICSI, PICSI and testicular spermatozoa-ICSI (T-ICSI) in patients with abnormal level of sperm DNA fragmentation (SDF). Summary answer Clinical effect of T-ICSI procedure exceeds the effect of PICSI and ICSI in male infertility with high level of SDF. What is known already The effectiveness of ART in male infertility depends on sperm chromatin structure integrity. High level of SDF may be reason of IVF/ICSI procedure failures. Several studies have demonstrated impaired clinical outcomes of ICSI in patients with abnormal level of SDF. SDF has a positive correlation with pregnancy loss. It has been shown the benefit of using of testicular spermatozoa in ICSI in high level of SDF. PICSI is a method of the selection of sperm that may reduce the risk of reproductive losses. The data on the comparison of the effectiveness of these methods in high SDF are limited. Study design, size, duration Data was collected from 2019 to 2021 years. The couples with male factor of infertility were enrolled in the study. All the patients had abnormal level of SDF (30-45%). The age of women was less than 35 year. The couples were randomly selected and divided in 3 groups according the ART method: I (n = 25) ICSI; II (n = 22) PICSI; III (n = 20) T-ICSI procedure. Participants/materials, setting, methods The trial involved 67 couples. The mean patients age was 33,3±6,2 y (24-46), the women age - 29,1±2,3 (25-33). All of the men were non-smokers, had idiopathic pathospermia, high level of SDF and underwent unsuccessful courses of empirical therapy in anamnesis. SDF level was measured by TUNEL method. 52 (77,6%) couples had unsuccessful attempts of ICSI including pregnancy loss. The fertilization, blastocyst formation (Gr A/B), clinical pregnancy and live birth rates were measured. Main results and the role of chance It was not significant difference in SDF level in the groups: I group - 37,3±3,1%; II group - 37,2±3,5%; III group - 37,8±3,2% (III). The fertilization rate was comparable in the groups I, II and III - 65,4%; 66,6%; 63,3% respectively. Blastocyst formation (good quality embryos) rate was 41,2%; 46,1%; 56,1%(p < 0001) in I, II and III groups respectively. Clinical pregnancy and life birth rates were 36,0% [9/25]; 40,9%[9/22] (p < 0,05); 45,0%[9/20] (p < 0001) and 20,0% [5/25]; 27,2%[6/22]; 35%[7/20] in ICSI, PICSI, T-ICSI groups respectively. It was significant difference in miscarriage rate 16%, 13,7% (p < 005), 10% (p < 0001) in I, II и III groups respectively. Limitations, reasons for caution T-ICSI might compensate for ART failure in high level of SDF cases. Testicular sperm retrival is an invasive intervention, and its benefit must be carefully justified in each cases. Further studies are needed to determine whether infertility with sperm DNA damage can be overcomed by less invasive methods than T-ICSI. Wider implications of the findings There was no significant difference in fertilization rate in ICSI, PICSI and T-ICSI. T-ICSI demonstrated the best blastocyst formation rate and clinical outcome in trial. According results of our investigation PICSI might increase pregnancy rates and reduce pregnancy losses when compared with ICSI in infertility in high SDF. Trial registration number not applicable
В обзоре приведен анализ отечественных и зарубежных литературных источников по изучению структуры хроматина сперматозоидов. Описаны патогенетические пути формирования ДНК-фрагментации гамет. Представлены данные о взаимосвязи между нарушениями целостности генома сперматозоидов c частотой наступления беременности и родов при лечении методами вспомогательных репродуктивных технологий, рисками привычных потерь беременности. Оценена диагностическая значимость фрагментации ДНК сперматозоидов при мужском факторе бесплодия. Ключевые слова: хроматин сперматозоидов; ДНК-фрагментация сперматозоидов; обзор литературы; мужское бесплодие; варикоцеле; окислительный стресс; реактивные формы кислорода; вспомогательные репродуктивные технологии; невынашивание беременности Финансирование. Исследование не имело спонсорской поддержки. Конфликт интересов. Авторы заявляют об отсутствии конфликта интересов. Вклад авторов: М.Н. Коршунов -обзор публикаций по теме статьи; Е.С. Коршунова -написание текста рукописи; П.С. Кызласов -обзор публикаций по теме статьи; Д.М. Коршунов -обзор публикаций по теме статьи; С.П. Даренков -обработка литературных данных.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2025 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.